Apply the Feature of Entropy Convergence of ACO to Short the Runtime of Gene Order

Alzheimer disease (AD) is the most common form of dementia. To find a way of cure it, gene study is necessary. And gene order is a new conception of gene study currently, where gene order refers to a permutation of genes in which similar genes are ordered together one by one, and optimal gene order can be abstracted as shortest TSP route. Currently only two types of tools are reported to calculate gene order, which are Genetic Algorithm (GA) and Ant Colony Optimization (ACO). In these two types, one bottleneck of computation is that their runtime is too long while gene data is too large. To weaken the bottleneck, in this paper, the feature of entropy convergence of ACO is used as the termination criterion of ACO to speed up the computation of AD gene order. Experiment shows that the method proposed in this paper has obvious advantage on runtime and solution quality.